S-CAS: Smart channel assignment scheme for wireless sensor networks

Abstract

This paper proposes a smart channel assignment scheme (S-CAS) for wireless sensor networks. S-CAS performs statistical analysis of load pattern of the individual cluster and adjusts channel assignment dynamically as per requirement generated by the current traffic load. It follows a static schedule during infrastructure establishment based on initial traffic conditions and previous history and later, it utilizes dynamic schedule for channel adjustments. Over a period of time, learning efficiency of the load prediction increases due to availability of more traffic patterns, in turn, reduces transmission delay and improves the system throughput and network life significantly. It provides seamless data transmission even in peak hours and is highly immune to random traffic change. This algorithm optimizes the channel utilization by best-of-four strategies and meets real time sensor network applications requirement. This algorithm is distributed at clusters level and central at node level i.e. it collectively possesses the advantages of all four: static, dynamic, central, and distributed channel assignment approaches. The simulation for 100 nodes network is done and results are discussed. This paper ends with planned future work on traffic prediction accuracy.